This essay examines what Skaldskaparmal can tell us about Snorri Sturluson's conceptualisation of the kenning and his justification of it as the primary mode of poetic diction within skaldic verse.  

Skaldskaparmal

Old Norse Supervision Essay - Supervision 3

Thomas Gizbert

12^th May 2009

Myrkt þykki mér þat mælt at kalla skáldskap með þessum heitum ('That seems to me an obscure way of speaking, to call poetry by these names').  

Consider Snorri's conceptualisation of the kenning and his justification of it as the primary mode of poetic diction within skaldic verse.  

Skaldic poetry, upon first glance, can be overwhelmingly confusing, due to its obtuse nature and the fact that the meanings of each verse are often hidden within kennings and other metaphorical literary techniques. In its interpretation, perhaps no medieval work is more important than Snorri Sturluson's Skáldskaparmál, written in the first half of the thirteenth century - a time when skaldic poetry was already going out of fashion - with the purpose of better enabling people to understand skaldic verse. Snorri achieves this by listing a great number of kennings on all of the popular subjects of the skalds of the Viking-age; gods, kings, battle, gold, men, women, nature, arms, and armour. The question which it is important to ask, then, in this context, is the extent to which Snorri's information can be trusted. He is far displaced in terms of both time and space from the skalds whose poetry he records, and many of the poems which he quotes as exemplars are not written down anywhere else, with centuries having passed between composition and being recorded. One way to determine how Snorri's account of these kennings might be distant from how they were perceived by their original composers would be to examine what we can find out about the way in which Snorri himself viewed the kenning, and its place within skaldic verse. In this essay, then, Snorri's treatment of the kenning shall be examined, and the themes which underlie his depiction of them will be analysed, as well as his justification of his elevation of the kennings to primacy among the modes of skaldic verse.

In order to consider Snorri's concept of the kenning, we must look no further than Snorra Edda itself. Snorri writes in Skáldskaparmál

'Þrenn erg rein skáldskaparmáls... Svá: at nefna hvern hlut sem heitir; Ç«nnur grein er sú er heitir fornÇ«fn; in ‏þriðja málsgrein er kÇ«lluð er kenning...'

                                                            -- Skáldskaparmál, 1

The first thing that we notice about kennings is that they are considered very highly by Snorri, who lists them as one of the three main types of skáldskaparmál (poetic language or diction). Indeed, in the Uppsala manuscript, there are only two categories, leading some scholars to believe that the idea of fornafn, ill-defined as it is, is a later addition to the work[1]. Snorri then dedicates most of Skáldskaparmál to the exemplification of kennings, both in the sense of how they work and what specific kennings mean. He goes on to write

'...ok er sú grein svá sett at ver kÇ«llum Óðin eða Þór eða T‎ý eða einnhvern af Ásum eða álfum, at hverr ‏þeira er ek nefni til, þá tek ek með heiti af eign annars Ássins eða get ek hans verka nokkvorra. Þá eignask hann nafnit en eigi hinn er nefndr var...'

                                                            -- Skáldskaparmál, 1

This, then, gives us a broad definition of Snorri's conceptualisation of kennings; a broad metaphor that can connect two separate ideas by adding a modifier to one - for example, Tyr becomes linked to Óðinn by adding the modifier 'victory', 'hanged', or 'cargo' ('svá sem vér kÇ«llum Sigt‎ý eða Hangat‎ý eða Farmat‎ý, þat er ‏þá Óðins heiti', Skáldskaparmál, 1). Likewise poetry is linked to Óðinn in the kenning 'hrostabrim AlfÇ«ðr' (Skáldskaparmál, 2) - the 'all-father' is a fornafn for Óðinn, and his 'malt-surf' is poetry.  

Even with his first, very specific description of a type of kenning, Snorri attaches the idea to mythology, focussing on names for the gods. This may be because he wishes to emphasise the ancient origin of this type of literary device, hearkening back to a pre-Christian past. Snorri goes on to state the importance of learning the mythological tales which provide the context for the kennings:

En ‏þetta er nú at segja ungum skáldum ‏þeim er girnask at nema mál skáldskapar ok heyja sér orðfjÇ«lða með fornum heitum eða girnask ‏þeir at kunna skilja ‏þat er hulit er kveðit... En ekki er at gleyma eða ósanna svá ‏þessar sÇ«gur at taka ór skáldskapum fornar kenningar þær er hÇ«fuðskáld hafa sér líka látit.

                                                           -- Skáldskaparmál, 1

It is clear from this that Snorri sees the kenning as, if not something directly related to Norse mythology, something which relies on knowledge of the underlying stories for much of its effect. Much has been written on the riddling nature of skaldic verse, with debate still raging about whether a poem's description would have been intelligible to its audience at once, or whether the effect was to confound the listener in a whirlwind of apparently unrelated terms which could be memorised and mulled over later so that the full meaning of the poem would be divulged[2]. It is probable that Snorri devised his work as a sort of key for the decipherment of Viking-Age skaldic verse, and not with the express intention that poets use it for the creation of their own works. In this case his focus on kennings linked to paganism, which might have been less easy to decipher than those which were simple metaphors, seems less of an imbalance than otherwise. For example, of the first twenty stanzas which Snorri uses as exemplars in Háttatal, only five of them have mythologically-based kennings in them. Of these, four only have one (two of which simply give King Hákon names of gods), and the fifth has four, all of which refer to land by calling it Óðinn's wife or beloved, or Þórr's mother. Comparing this to the much more numerous examples of kennings which are not related to Norse mythology, we can see that the former comprise a definite minority. The same pattern occurs in Skáldskaparmál, in which, although kennings referring to the gods or referencing mythological events take pride of place, the majority of the work is dedicated to metaphors for battle, men, women, nature, arms, and gold. The passage quoted above, then, points to a dual purpose in Skáldskaparmál: firstly to list kennings used in skaldic poetry, and secondly to elucidate on where they might have originated.

Another facet to Snorri's conceptualisation of the kenning is the manner in which they are described as secretive or riddling. In the section quoted above, Snorri talks about '‏þat er hulit er kvedit' (Skáldskaparmál, 1) - literally, that which is expressed in a hidden fashion. Previously in Skáldskaparmál, Ægir talks to Bragi about what he thinks of the kennings for gold which Bragi has mentioned, and what he says about them could be a reflection of Snorri's thoughts on the subject:

            '‏Þat ‏þykki mér vera vel fólgit í rúnum.'

                                                                        -- Skáldskaparmál, G56

This idea about kennings as being the 'hiding' of meaning in 'language' (interestingly, 'rúnar' also has the meaning of 'secret lore' or 'hidden wisdom', an idea which might link back to the riddling and obtuse nature of skaldic poetry) makes one think back to the idea of the riddling nature of skaldic verse, and also to the magical power of language in Norse paganism, in which Óðinn hung himself from Yggdrasill for nine days in return for knowledge of the runes[3], and can summon people from the dead in return for knowledge[4]. The linking of these ideas is an interesting one, as it connects skaldic poetry to a form of pagan magic, and the knowledge of kennings to something divine. The phrasing of this extract is also worth commenting upon. The adverb 'vel' has in Old Norse the dual meaning that 'well' has in English - both indicating 'fully, thoroughly' and 'for good purpose, with good outcome' - giving the sense that Ægir, and possibly Snorri also, admires the fact that, or the adroitness with which, the meaning has been hidden in the kenning. In all, this sentence indicates a high regard of the art form of the kenning, and an appreciation of the origin of the technique in pagan mythological times, in which secret language played a large part.

Another thing to keep in mind when considering Snorri's conceptualization of the kenning is the various 'rules' which Snorri imposes upon the creators of kennings. He writes in Skáldskaparmál that the composition of new kennings or extension of meaning of ancient ones is acceptable if it is 'natural'.

            Því er ‏þat kallat nýgervingar alt er út er sett heiti lengra en fyrr finnsk, ok ‏           þykkir ‏þat vel alt er með líkindum ferr ok eðli

                                                                        -- Skáldskaparmál, 33

The words 'líkindum', meaning 'probable', and 'eðli', meaning 'natural', seem, however, fairly arbitrary, and perhaps show the limits of Snorri's attempts to impose order on a system which is not inherently orderly (it can be argued that the same thing happens when he attempts to provide an ordered account of Norse mythology: discrepancies and vaguities appear in the light of his critical analysis). This same arbitrariness appears later in the work.

Þessi nÇ«fn himins eru rituð, en eigi hÇ«fum vér fundit í kvæðum Ç«ll ‏þessi heiti. En ‏þessi skáldskaparheiti sem Ç«nnur ‏þykki mér óskylt at hafa í skáldskap nema áðr finni hann í werka hÇ«fuðskálda ‏þvílík heiti.

                                                            -- Skáldskaparmál, 56

The word 'óskylt' in this context means something like 'improper', again hinting at specific guidelines for the composition on kennings which Snorri is not able to provide. His judgement on whether or not they are 'proper' seems to be based upon whether they are similar '‏þvílík' to those of Viking-Age skalds. This traditionalism which is inherent in Snorri's work will be examined in more detail later in the essay. In other instances, he seems to attempt to impose rules on specific usages of his general kennings, rules which again seem arbitrary because they attempt to legislate what might be seen as common sense. An example of this is

Vargr heitir d‎ýr. Þat er rétt at kenna við blóð eða hræ svá at kalla verð hans eða drykk. Eigi er rétt at kenna svá við fleiri d‎ýr.

                                                            -- Skáldskaparmál, 58

Surely it is not strictly necessary to say that, although the poet can talk about the wolf in terms of blood-drinking and corpse-eating, it is not correct ('rétt') to do the same in reference to, say, a cat or a horse. Snorri in this instance is probably not insulting the reader's intelligence, but instead attempting to impose order upon a system of poetry, which, though a very complex subject, is fluid, based on the natural world, rather than rigid and categorical.

The question, however, of Snorri's justification for the kenning being the primary mode of diction in skaldic verse is a more difficult one. Indeed, he often alludes to kennings as being an important part of the arsenal of skalds, but he does not overtly say why this is. In Skáldskaparmál he writes, about the stories of Norse mythology:

             Eptir ‏þessum sÇ«gum hafa flest skáld ort ok tekit ymsa ‏þáttu.

                                                                        -- Skáldskaparmál, 42

Here 'flest' is the superlative of 'fleiri' meaning 'many'; 'flest' therefore means 'most', and Faulkes glosses it as 'almost all'[5]. The claim, then, is that most or almost all skalds have taken '‏various strands' ('ymsa þáttu') from the mythological stories, putting forward the view that the kennings based on Norse paganism are prevalent within the Viking Age skaldic corpus, and that kennings themselves are even more so. Later in the text, he writes, regarding kennings for weapons and armour, that:

Ǫllum ‏þessum kenningum er marga lund breytt þvíat ‏þat er flest ort í lofkvæðum er ‏þessar kenningar þarf við.

                                                            -- Skáldskaparmál, 49

Again, this indicates that kennings are very common in praise poetry in particular and in skaldic poetry in general. The operative word here is '‏þarf' which has the meaning 'needing' or 'in want of'. It has the sense of 'necessary' and Faulkes renders it as 'required' in his translation of Snorra Edda[6]. This seems to be a fairly clear-cut example of the importance of kennings in skaldic poetry, even if it has the same abrupt arbitrariness as some of Snorri's other rules. Indeed, Snorri even gives examples in Háttatal of stanzas without kennings or metaphorical language at all, even though these seem to be the exception to the rule. Nevertheless, Snorri's language hints at an overall tradition that kennings are the norm in skaldic poetry, even if there was, before Snorri, no explicit statement of the fact, just as there was likely no written rule that one could only refer to wolves when using the phrase 'corpse-eating beasts'.

It seems that one of Snorri's justifications of the kenning as the most important mode of diction within skaldic verse can be found in the weight which he places on tradition. We have come across the theme of tradition before, both in Snorri's connection of kennings to Viking-age ideas about language and magic, and in his consideration of rules about kennings, most of which are based upon the authoritative weight of the pre-Christian skalds, and it can be argued that tradition is an underlying theme of Snorri's work. By the thirteenth century, when Snorri was writing, the heyday of skaldic poetry and praise poetry was long over, and the popularity of the art form would continue to decline to nothing over the next two hundred years. It is perhaps more fitting to describe Snorri's Skáldskaparmál and Háttatal as histories of skaldic verse, rather than manuals for their composition. All that which Snorri writes, he justifies using the examples of Viking-age skalds.

Enn skal láta heyra dœmin hvernig hÇ«fuðskáldin hafa látit sér sóma at ‎ýrka eptir ‏þessum heitum ok kenningum...

                                                            -- Skáldskaparmál, 2

The word 'hÇ«fuðskáldin' in this context - literally 'head-poets' - means something like 'great poets', and shows the extent to which Snorri considers the work of the poets of pagan-era Scandinavia more valid than those of the present day. His conservative ideas of poetry show through in this extract, in which he implies that one of the main justifications of the usage of these kennings is the fact that these 'great poets' found it 'sóma' - 'fitting', 'seemly' - to use them. The extent to which he makes the effort to justify each of his examples with a quotation from a Viking-age skald, and to proffer a caveat if he cannot, shows that he considers their usage of these 'hÇ«fuðskáldin' to be both a verification of their position as the primary form of poetic diction and also the justification of the same.

It would be wrong, however, to discount artistic proficiency as a reason for Snorri's conceptualisation of the kenning as the primary form of poetic diction. Although various critics believe due to his depiction of the creation of kennings that he considers them more of a craft than an art, there are episodes in Skáldskaparmál in which he almost revels in the artistic possibilities provided by kennings, and the varied effects which they can create as a literary technique. Snorri writes, while listing ways to describe men:

Hvernig skal kenna mann? Hann skal kenna við verk sín, ‏þer er hann veitir eða þiggr eða gerir... Mann er ok rétt at kenna til allra Ása heita. Kent er ok við jÇ«tna heiti, ok er ‏þat flest háð eða lastmæli. Vel ‏þykkir kent til álfa.

                                                            -- Skáldskaparmál, 31

This extract provides a rare glimpse into Snorri's interest in the effects that these kennings might have on the audience of the poem. Again, what is said - that calling someone the name of a giant is criticism ('lastmæli' - 'blame-speech', 'slander'), while calling them by the name of an elf is complimentary ('vel' - 'well', 'kind') - might have been fairly obvious to any contemporary who had been paying attention to the stories which he had recorded before this point in Skáldskaparmál, but it is nevertheless interesting that he says it, giving the reader confirmation that he is aware of the artistic effect which kennings can have on skaldic poetry, not only confusing the poem and making it more difficult to decipher, but also linking ideas in the minds of the audience. This passage and others like these are confirmation that Snorri sees kennings as not only worthwhile due to tradition, but also as a literary technique.

Another point in Skáldskaparmál at which Snorri's appreciation of kennings beyond their heritage comes to the forefront is at the very end of the prose section, in which he describes the techniques and effects of 'word play'.

Læti er tvent. Læti heitir rÇ«dd, læti heitir œði, ok œði er ok ólund. Reiði er ok tvíkent. Reiði heitir ‏þat er maðr er í illum hug, reiði heitir ok fargervi skips eða hross... Þvílík orðtÇ«k hafa menn mjÇ«k til ‏þess at yrkja fólgit ok er þat kallat mjók ofljóst.

                                                            -- Skáldskaparmál, 74

The name which Snorri gives to this technique is 'ofljóst' (literally 'too-clear') which is glossed by Faulkes as 'excessively clear' or 'obvious'[7]. Although this does not fall under the category of kenning per se, the idea of a metaphorical or dual-faceted phrase which can be used to confuse or conceal meaning is in many ways similar to a kenning, and the description of its effect on the audience could in many ways be applied to kennings.

Þessar greinir má setja svá í skáldskap at gera ofljóst at vant er at skilja ef aðra skal hafa greinina en áðr ‏þykki til horfa in fyrri vísuorð.

                                                            -- Skáldskaparmál, 74

It seems from this that Snorri is delighting in playing tricks with his audience. The fact that he explains in some depth what this particular technique can bring to a stanza of skaldic poetry, whereas he does not do so with many of the other techniques that he has explained, gives the reader the feeling that he is particularly interested in the ways in which skalds can play with their audience's expectations through language. Also the way in which he implies that it is desirous to make the poems difficult or perplexing could perhaps indicate that this is one of the main goals when it comes to the inclusion of kennings in skaldic poetry. In Skáldskaparmál, when Snorri talks about kennings, words like 'fólgit' ('hidden', G56), 'myrkt' ('obscure', G58), and 'hulit' ('secretly', 1) are rife. From this, and from what we have seen above regarding the ideas of the riddling nature of skaldic poetry, we can guess that one of Snorri's justifications in using kennings as the primary mode of poetic diction is the fact that it can be made to obscure meaning so effectively. Considering this along with the previous point - that kennings can give nuances of meaning to skaldic poetry - it is possible to see the appeal which this technique must have held for the skald, both hiding the true sense of the poem and revealing other layers of meaning through connotation and metaphor.

There are, however, points in Skáldskaparmál in which Snorri criticises the failings of the kenning. He writes towards the end of the work, during his list of kennings for Christ and worldly kings

Þar koma saman kenningar, ok verðr sá at skilja af stoð, er ræðr skáldskapin, um hvárn kveðit er konuginn...

                                                            -- Skáldskaparmál, 53

Usually the word 'saman' is translated as 'together', but it has an alternate meaning of 'in common' or 'homonymic' (Faulkes translates it as 'ambiguous'[8]), and it is used in the latter sense later in Skáldskaparmál when Snorri is talking about 'word-play' - 'Slíkt sama eru ok Ç«nnur mÇ«rg nÇ«fn ‏þau er saman eigu heitit margir hlutir', (Skáldskaparmál, 74). It seems as if in this instance Snorri is offering a warning to readers or other poets, that if the kenning is ambiguous then the context should provide clues about who or what is being referred to. This should probably not be seen as a disparaging comment on the usefulness of kennings, especially in light of what we have already seen regarding the desirability of dual meanings and ambiguity in skaldic poetry. Instead, it can be interpreted as an admission of the imperfection of kennings as a mode of poetic diction, and an explanation of why other forms are also required.  

Now that Snorri's conceptualisation of the kenning and his justification of it as the primary mode of poetic diction in skaldic poetry have been examined, there remain a few notes which must be made regarding the way in which Snorri characterises the kennings. First, there has been a broad scholarly consensus in the latter half of the twentieth century that many of what Snorri refers to as kennings, in that their names have been substituted for a term related through stories or otherwise, should be taken to be metaphors. Faulkes writes in the introduction to his edition of Skáldskaparmál that 'even kennings which seem to us obviously metaphorical, such as when gold is called fire of the sea... are explained by Snorri in terms of substitutions... That is, a word for sea may be substituted for the name Ægir as a variation on the kenning-type 'fire of Ægir', based according to Snorri on the story of how Ægir used gold as a source of light when he entertained the Æsir to a feast.'[9] We can see also that Snorri is aware of the extent of these substitutions in skaldic poetry, but he attributes their existence to extension of meaning over time, and warping of tradition due to 'recent poets' ('yngri skáld'):

En ‏þessi heiti hafa svá farit sem Ç«nnur ok kenningar, at hin yngri skáld hafa ort eptir dœmum henna gÇ«mlu skálda, svá sem stóð í‏ þeira kvæðum, en sett síðan út í hálfur ‏þær er ‏þeim þóttu líkar við ‏þat er furr var ort, svá sem vatnit er sænum en áin vatninu en lœkr ánni.

                                                            -- Skáldskaparmál, 33

This complex explanation is one which is necessitated due to Snorri's emphasis on the more valid traditionalism of the Viking-age skalds, and his emphasis on straight substitution over creative interpretation and metaphor, but his explanation still smacks of an attempted rationalisation based on false premises. Faulkes describes Snorri's way of thinking in regard to the literary techniques of skaldic poetry in that 'he sees poetical language largely in terms of substitutions of one name for another, rather than in terms of transference of meaning.'[10] In studying Snorri's characterization of kennings, therefore, we must be aware of the fact that many of the explanations that he gives for them might be, to a large extent, tainted by the way in which he sees them more as products of literal substitution, less an artistic linking of ideas.

Following on from this, the second point which must qualify our view generally on Snorri's characterization of skaldic poetry is that it is possible that he made up mythological stories in order to account for the metaphors which he interpreted as kennings. Roberta Frank argues this in some depth in her article 'Snorri and the mead of poetry', in which she writes that 'Snorri cites a number of tenth-century stanzas to illustrate his conclusions regarding the poetry kennings of the skalds, kennings for whose explication he built his mythological edifice. If it can be shown that he misread those stanzas, his edifice topples, taking with it the idea that centuries of oral tradition underlie all portions of the myth as he tells it.'[11] Her work comes to interesting conclusions, in line with what we already know about Snorri's view of the weight of the Viking-age tradition behind the skaldic poetry and his willingness to go to great lengths to rationalise large amounts of kennings as being substitutions based on mythological stories rather than metaphorical language. Frank's paper makes the point well that the reader of Skáldskaparmál must be aware that Snorri's views of the kenning may not have conformed to the views of the Viking-age poets, and that a lot of what he says may be tainted as a result.

In conclusion, the underlying theme of the traditional nature of kennings in skaldic poetry, an idea which Snorri adheres to solemnly and one which pervades his work, both validates and limits Skáldskaparmál. Snorri begins with a relatively firm idea of the definition of the kenning, but he then struggles with knowing what to classify as a kenning and what not to. From the outset he sees kennings as being linked to Scandinavian pre-Christian mythology, and although this lends great value to his work, it also might often mean that he finds himself in a situation where he must attempt to work backwards in order to extrapolate a myth he assumes lost from the extent kennings which he believes relate to it. While there are points in Skáldskaparmál in which the reader can see that Snorri admires the way in which the skalds have used kennings to hide meaning, but add layers of sense, the tradition-based approach which he employs diminishes the possibility of viewing the kennings as a form of creative expression. Likewise, the strength lent to the work by Snorri's categorical invocation of the Viking-age skaldic corpus which is always at hand to back up whatever he proposes regarding kennings is offset by his imposition of vague or common-sense rules on the art form. In all, although Skáldskaparmál is invaluable to the student of skaldic poetry and Scandinavian pagan mythology, it is nevertheless necessary to be aware of the way in which Snorri's conceptualisations and justifications of kennings as the primary mode of poetic diction can be detrimental to forming a clear picture of the way in which Viking-age skaldic poetry worked, and of pre-Christian Scandinavian mythic tradition.

Bibliography:

Clunies Ross, M., A History of Old Norse Poetry and Poetics (Cambridge, 2005)

Clunies Ross, M., Skáldskaparmál: Snorri Sturluson's ars poetica and medieval

theories of language (Odense, 1987)

Frank, R., Old Norse Court Poetry: The Dróttkvætt Stanza (London, 1978)

Frank, R., 'Snorri and the mead of poetry' in Speculum Norroenum: Norse Studies in

Memory of Gabriel Turville-Petre, ed. U. Dronke et al. (Odense, 1981)

pp. 155-70  

Larrington, C., trans., The Poetic Edda (Oxford, 1996)

Nordal, G., Tools of Literacy: The Role of Skaldic Verse in Icelandic Textual Culture

of the Twelfth and Thirteenth Centuries (Toronto, 2001)

Snorri Sturluson, Edda, trans. A. Faulkes (London, 1987)

Snorri Sturluson, Edda: Prologue and Gylfaginning, ed. A. Faulkes, 2^nd ed. (London,

2005)

Snorri Sturluson, Edda: Skáldskaparmál - 1: Introduction, Text and Notes, ed. A.

Faulkes (London, 1998)

Snorri Sturluson, Edda: Skáldskaparmál - 2: Glossary and Index of Names, ed. A.

Faulkes (London, 1998)

Snorri Sturluson, Edda: Háttatal, ed. A. Faulkes (London, 1991)

Zoëga, G. T., ed., A Concise Dictionary of Old Icelandic (London, 2004)

[1] Faulkes, Skáldskaparmál, pp. xxxiv-xxxv

[2] Franks, Old Norse Court Poetry, p. 28

[3] 'Sayings of the High One', Poetic Edda, szs. 138-9

[4] 'Baldr's Dreams', Poetic Edda, sz. 4

[5] Faulkes, Skáldskaparmál, p. 275

[6] Snorri Sturluson, Edda, p. 119

[7] Faulkes, Skáldskaparmál, p. 366

[8] Snorri Sturluson, Edda, p. 127

[9] Faulkes, Skáldskaparmál, pp. xxx-xxxi

[10] ibid, p. xxx

[11] Frank, 'Snorri and the mead of poetry', p. 157